Indoor nav, including using MEMS sensors, is already taking off. There are over a hundred companies bringing indoor nav to market http://goo.gl/NpA4Vk and virtually all of them are incorporating, or working now to incorporate, use of sensors. This is beacuse radio-based techniques are not precise enough in handling small movements or in handling transitions smoothly. Also, as E911 legislation warms up, barometers for altitude will become increasingly important.
You folks who want indoor location to help you find stuff: do you think the RETAILERS know where it should be, or or if there is any in stock? That makes me laugh! More often then not, I check stock on many large retailers for stock status of an item in nearby stores, and 80% of the time the information is WRONG, either the item shown out of nstock has lots on the shelves, or items shown as in-stock may not even be carried in that particular store! Happens with Home Depot, Walmart, Sears, etc. Even the HEADQUARTERS "showplace" Home Depot store I usually patronize suffers from this. I've been told by store and department managers that part of the problem is they ship stuff to stiores from a centralized regional warehouse in a 53-foot trailer. Their computers show the items in stock as soon as the trailer is delivered to the parking lot; it may take days or even weeks to get put on the shelves (and then not always where they should be). Knowing where YOU are is the EASY part!
Regarding indoor vs. outdoor nav and the need, or lack thereof, for pressure sensors -- the one app I have found useful that provides altitude data is one that is used primarily for skiing & snowboarding. It uses GPS to locate the user's x,y and z, and of course is intended only for outdoor use, but does quite a good job of tracking your altitude changes throughout your ski day and easily distinguishes a lift ride up the mountain from a ski run down.
I can't think of many scenarios in which a user would care to know his or her altitude within a building or to know which floor they're on. That is generally something people can keep track of on their own!
@Peter: Indoor navigation is very desirable solution in many situations. But its solution, both hardware and software algorithms, may be very demanding.
As for Ramo, I would say we expect solution from him and his organization. He should not complain like end user . They make mobile phones, they need market and for that they need to innovate and work hard. Not complain and not act like end user.
I read somewhere that Google uses this technique of doing indoor navigation of buildings and malls by using the known location of all the various WiFi antennas. It is similar to the way that they can do navigation by knowing where the cell stations are.
Assuming there are separate antennas upstairs and downstairs, you wouldn't need a separate altitude (pressure) sensor.
I've also heard that mall owners plan to use WFi information along with cell phone ID numbers to figure out people's shopping habits.
As I've mentioned elsewhere, the pressure sensor data may be skewed by HVAC (climate control) induced building pressure differentials and by weather induced air pressure changes. At the same time, there is a very real interest by consumers in being able to navigate within large malls and buildings. The building owners also have an interest in people being able to find their way around. A combination of beacons and WiFi transmitters in known locations should enable map programs to determine the geographical location and elevation (floor) of a smartphone. People walk much slower indoors than they drive outdoors so some parts of the problem are simplified. An alternative approach is the use of images taken by the phone to locate the user and then use inertial navigation from there.
The reason indoor navigation has not taken off is due to the fact that the consumer grade MEMS sensors (gyro and accels) that are currently being put into cell phones today are nowhere near good enough (performance wise) for this type of application.
Consumer grade sensors have too much drift over time to be accurate enough for indoor navigation, hence the statement in the article "often showed the users being outside the mall when they were inside". The sensor drift causes error in the position and this error adds up very fast vs time. So with consumer grade sensors within 30 seconds you could be 100s - 1000s of meters off from where your phone says you are and where you really are.
I believe in order to make indoor navigation viable with consumer grade sensors another reference will be needed to help correct for the sensor drift. I know there are various reference schemes that are being investigated such as WiFi and optical to help correct for the drift. In order to have a succesful indoor navigation experience either the sensors will have to get a LOT better or some reference will also be needed in the sensor fusion to correct for the various sensor drifts.